learning_rate: 2e-05
train_batch_size: 4
eval_batch_size: 4
seed: 42
optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: cosine_with_restarts
lr_scheduler_warmup_steps: 102
num_epochs: 10

Training Loss	Epoch	Step	Accuracy	F1	F1 Class 0	F1 Class 1	F1 Class 2	F1 Class 3	F1 Class 4	Validation Loss	Precision	Recall
1.6069	1.0	102	0.4706	0.3640	0.2857	0.0	0.0	0.24	0.6406	1.4170	0.4690	0.4706
1.4203	2.0	204	0.4706	0.3720	0.4444	0.0	0.0	0.0	0.6496	1.4936	0.3207	0.4706
1.2376	3.0	306	0.4314	0.3672	0.4935	0.0	0.0	0.0	0.6098	1.4866	0.3417	0.4314
1.011	4.0	408	0.5196	0.4408	0.5532	0.0	0.0	0.1905	0.6909	1.5160	0.4023	0.5196
0.9701	5.0	510	0.5	0.4395	0.5556	0.0	0.0	0.25	0.6667	1.7326	0.3972	0.5
0.8141	6.0	612	0.5490	0.4846	0.6190	0.0	0.125	0.2727	0.6964	2.2498	0.4802	0.5490
0.5545	7.0	714	0.5098	0.4677	0.5532	0.1176	0.1333	0.2609	0.6667	2.7445	0.4653	0.5098
0.4284	8.0	816	0.5294	0.5048	0.5333	0.2105	0.3158	0.2609	0.6939	2.7553	0.5024	0.5294
0.3441	9.0	918	0.5294	0.5118	0.5333	0.2	0.3810	0.2609	0.6947	2.7991	0.5085	0.5294
0.2863	10.0	1020	0.5294	0.5057	0.5333	0.1111	0.3810	0.2609	0.7010	2.8321	0.4993	0.5294

matiashoyl/modernbert-match-user-32329

modernbert-match-user-32329